We analyse the dynamics of convolutional filters' parameters of a convolutional neural networks during and after training, via a thermodynamic analogy which allows for a sound definition of temperature. We show that removing high temperature filters has a minor effect on the performance of the model, while removing low temperature filters influences majorly both accuracy and loss decay. This result could be exploited to implement a temperature-based pruning technique for the filters and to determine efficiently the crucial filters for an effective learning.

Thermodynamics modeling of deep learning systems for a temperature based filter pruning technique / Lapenna, M; Faglioni, F; Fioresi, R. - In: FRONTIERS IN PHYSICS. - ISSN 2296-424X. - 11:(2023), pp. 1145156-1145162. [10.3389/fphy.2023.1145156]

Thermodynamics modeling of deep learning systems for a temperature based filter pruning technique

Lapenna, M;Faglioni, F;
2023

Abstract

We analyse the dynamics of convolutional filters' parameters of a convolutional neural networks during and after training, via a thermodynamic analogy which allows for a sound definition of temperature. We show that removing high temperature filters has a minor effect on the performance of the model, while removing low temperature filters influences majorly both accuracy and loss decay. This result could be exploited to implement a temperature-based pruning technique for the filters and to determine efficiently the crucial filters for an effective learning.
2023
17-mag-2023
11
1145156
1145162
Thermodynamics modeling of deep learning systems for a temperature based filter pruning technique / Lapenna, M; Faglioni, F; Fioresi, R. - In: FRONTIERS IN PHYSICS. - ISSN 2296-424X. - 11:(2023), pp. 1145156-1145162. [10.3389/fphy.2023.1145156]
Lapenna, M; Faglioni, F; Fioresi, R
File in questo prodotto:
File Dimensione Formato  
fphy-11-1145156.pdf

Open access

Tipologia: Versione pubblicata dall'editore
Dimensione 2.78 MB
Formato Adobe PDF
2.78 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1320786
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
social impact